The Agent Skills Directory

[COMMAND_EXECUTION]: The visual generation script scripts/generate_market_visuals.py uses subprocess.run to coordinate tasks between the market report skill and other specialized visualization skills (scientific-schematics and generate-image). This implementation is secure as it uses list-based command construction and does not invoke a shell, effectively mitigating command injection risks.
[SAFE]: The skill processes external data gathered from the research-lookup tool to populate its 50+ page reports. While this creates a potential surface for indirect prompt injection (Ingestion: research tool outputs; Boundary markers: none; Capability: local subprocess execution; Sanitization: none), the implementation is consistent with the primary purpose of the skill and relies on structured LaTeX templates, posing a negligible risk for typical usage scenarios.

market-research-reports